Re: Help with 7 Circuit Project?

[nobody@xxxxxxxxxxxx (Dave Houston)]

It's one thing to train others in the care and feeding of a proprietary
system, it's another to understand the fundamentals of powerline
communications.

The fact that you had both the X-10 frequency and voltage levels wrong and
the fact that your knpwledge of UPB appears to be based on PCS press
releases indicate that your grasp of the fundamentals is weak.

There is an "oscilloscope" screenshot of the UPB pulse on pp8-10 at...

    http://www.ieee-denver.org/comsoc/presentations/UPB.pdf

You will find there are numerous things (e.g. triacs, ballasts, PF
capacitors) that will put similar noise pulses on the powerline. So it's
doubtful that the shape, position and frequency content of the pulse are
major factors in the robustness of UPB. The fact that they use a two-way
error correcting protocol probably has a far greater impact.

HAI recommends a phase coupler for all UPB installations. Why?

As I recall, UPB has a fairly low bitrate so claims that it is fast should
be taken with a grain of salt.

I may be wrong but my interpretation of Cenelec rules would mean UPB is
illegal in Europe. (5V maximum signal strength.)

Any talk of security for powerline or RF communications is ridiculous.
Unless the system uses rolling codes it is a simple matter to capture codes
and play them back - no decrypting is necessary.

That powerline communications can be robust is proven by the widespread
adoption of HomePlug's broadband over powerline. PLC broadband modules are
retailing in the $20 range. Should someone decide to use HompePlug for
lights and appliances, it's likely to be "lights out" for PCS and X-10.

UPB may work as advertised but, so far, most of the reports here have come
from dealers. There has been near zero feedback from end users. Given the
high cost, it's likely to popular only with installers (if it is as robust
as claimed). If the UPB computer interface could also do X-10, it would help
with market penetration.

Insteon (from SmartHome) would seem to win the published specifications
duel. It has a high bitrate (i.e. it's fast), uses a two-way
error-correcting protocol, is backwards compatible with X-10, uses both PLC
and RF, fills the network with a 2.5V signal level (each module functions as
a repeater), and doesn't cost much more than X-10. It would appear to comply
with Cenelec.

Again, there have been near zero end user reports on Insteon Given that you
can buy a 5 module Insteon starter kit for the cost of one UPB module
probably means we'll soon see end user reports.

BTW, long runs of copper, in and of themselves, have a minimal effect of
signal propagation. Noise, impedance mismatches and loading are the
overwhelming cause of low signal strength.

"Brett Griffin" <brett@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:

>> 5 years? It was first announced about 2.5 years ago, first shown about a
>> year ago, and has been shipping for only 9 months or so.
>
>Actally, PCS Lighting (who developed UPB technology) bagan alpha testing in
>1999, and beta testing in 2003.  Actually they are still testing (because
>Marshall, the engineer that discovered the technology in his garage, wants
>to be sure there are no errors like X-10 has.)   Products bagan shipping
>from PCS lighting in November of 2003, which is now 17 months according to
>my watch.
>
>> You are full of it. The loss of signal strength is independent of
>> distance.
>> There have been numerous studies that confirm this. It's the inductive and
>> capacitive loading that attenuates the signal.
>
>What do you think long runs of copper create?  At higher frequencies it
>attenuates the signal.  I am a little light on this, I admit.  However I
>know that long runs of cable translates into needing more signal strength.
>Several manufactures have explained this to me as well.
>
>> Most X-10 transmitters actually transmit 10Vpp at 120kHz. Noise is lowest
>> at
>> zero crossing.
>
>I am not looking to argue symantics here.  The point I am trying to bring up
>is that most manufacturers discuss a 5 volt signal and that is small enough
>for noise to mask the signal.  UPB does not have this problem as it
>transmits 40 volts above/below the sine wave.
>
>> Your description of a UPB "signal" sounds very much like noise pulses
>> which
>> are quite common from triacs, motors, ballasts, etc. As I understand it,
>> UPB
>> generates their pulse by rapidly discharging a capacitor. That sounds like
>> a
>> noise pulse.
>
>Actually technically speaking it is Pulse Frequency modulation if you desire
>to get specific, I was trying to keep the discussion general enough so the
>entire audiance would understand.  It may very well be as you describe a
>noise pulse.  However it is the timing and positioning of the signal that
>encodes the data.
>
>> I suggest you learn a few fundamentals before trying to pass yourself off
>> as
>> an expert. Spouting nonsense like this on topics you obviously do not
>> understand will get you an idiot label rather quickly.
>
>I know my fundamentals, and I know this topic very well, I have taught over
>a thousand dealers about this topic and have been involved in close to 100
>installs.  I was an HAI trainer until January 2005 when I left to be an
>independant consultant.  I also know the developers of UPB personally and
>speak with them frequently, I have the knowledge from the horses mouth, so
>to speak.  I am not spuoting any nonsense, as you chose to say.
>
>My point was to stress to those who have written off power line controls,
>they should re-consider.  UPB is a very reliable technology.  The experience
>I have with, shows that it works in any home with neutral. It is as reliable
>as a hard wired system, with out the cost.  With that being said it is a
>common agreement amoung the manufacturers that support UPB that in high
>density dwellings like Apartment buildings and condos, a hard wired system
>is a better choice.  In some cases it is more reliable than a hard wired
>system, for example Vantage has at least one circuit board in there that if
>it goes out, you have no lights.  UPB does not have any single point of
>failure, except for a black-out...